In this paper, counting objects techniques are proposed for fast pattern matching algorithm based on normalized cross correlation and convolution technique which are widely used in image processing application. Pattern matching can be used to recognize and/or locate specific objects in an image. It is one of the emerging areas in computational object counting. In this paper, introduces a new pattern matching technique called convolution based on pattern matching algorithm. Many different pattern matching techniques have been developed but more efficient and robust methods are needed. The pattern matching algorithm is used to identify the patterns similar present in image. With the patterns, identify the similarity measures of the given pattern to count the object present in the given image. An experimental evaluation is carried out to estimate the performance of the proposed efficient pattern matching algorithm for remote sensing as well as common images in terms of estimation of execution times, efficiency and compared the results with an existing conventional methods.
A very important feature extraction method that is commonly used in computer visions and image processing applications is counting of objects. This paper represents a modified sequential region labeling algorithm which counts the homogeneous region of different objects on image. It is based on 4-connected, 6-connected and 8-connected component technique. These algorithms scan the image pixel by pixel from left to right and top to bottom sequentially and assign a label to every foreground pixels in binary image. Salt and pepper noise is usually prevalent in such images. Removing this noise is an important issue. We propose median filter algorithm to removed such type of noise and obtain better results. This technique may be applied to uniform, nonuniform, regular, irregular objects with different shape, size and file formats. Binary images are obtained from color or greyscale images by proper thresholding. In this proposed method, the regions of various objects are found by region labelling process. These distinct regions are given the number of objects that are present inside the image. This algorithm is implemented on the .net technology. These methods produced good performance in term of accuracy. This is a process oriented task. So the machine having higher processing speed can serve the purpose better.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.